BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Melbourne
X-LIC-LOCATION:Australia/Melbourne
BEGIN:DAYLIGHT
TZOFFSETFROM:+1000
TZOFFSETTO:+1100
TZNAME:AEDT
DTSTART:19721003T020000
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=1SU
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:19721003T020000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260114T163642Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231214T144500
DTEND;TZID=Australia/Melbourne:20231214T150000
UID:siggraphasia_SIGGRAPH Asia 2023_sess151_papers_776@linklings.com
SUMMARY:Fluid Simulation on Neural Flow Maps
DESCRIPTION:Yitong Deng (Dartmouth College), Hong-Xing Yu (Stanford Univer
 sity), Diyang Zhang (Dartmouth College), Jiajun Wu (Stanford University), 
 and Bo Zhu (Dartmouth College)\n\nWe introduce Neural Flow Maps, a novel s
 imulation method bridging the emerging paradigm of implicit neural represe
 ntations with fluid simulation based on the theory of flow maps, to achiev
 e state-of-the-art simulation of inviscid fluid phenomena. We devise a nov
 el hybrid neural field representation, Spatially-sparse Neural Fields (SNF
 ), which fuses small neural networks with a pyramid of overlapping, multi-
 resolution, and spatially-sparse grids, that compactly represents long-ter
 m spatiotemporal velocity fields at high precision. With this neural veloc
 ity buffer at hand, we compute long-term, bidirectional flow maps and thei
 r Jacobians in a mechanistically symmetric manner, to facilitate drastic a
 ccuracy improvement over existing solutions. These long-range, bidirection
 al flow maps enable high advection accuracy with low dissipation, which in
  turn facilitates high-fidelity incompressible flow simulations that manif
 est intricate vortical structures. We demonstrate the efficacy of our neur
 al fluid simulation in a variety of challenging simulation scenarios, incl
 uding leapfrogging vortices, colliding vortices, vortex reconnections, as 
 well as vortex generation from moving obstacles and density differences. O
 ur examples show increased performance over existing methods in terms of e
 nergy conservation, visual complexity, adherence to experimental observati
 ons, and preservation of detailed vortical structures.\n\nRegistration Cat
 egory: Full Access\n\nSession Chair: Tao Du (Tsinghua University, Shanghai
  Qi Zhi Institute)\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_776&sess=sess151
END:VEVENT
END:VCALENDAR